What are we measuring when we measure task switch costs?

Task switching procedures are widely used to assess the processes supporting executive function and cognitive control, but there is wide variation in what subjects are required to do during these procedures and a lack of consensus about what, exactly, constitutes a task switch. The methodological variation in task switching experiments has revealed diverse factors that affect the magnitude of switch costs, thereby enriching our understanding of cognitive control. For example, the empirical phenomena uncovered by these procedures indicate that specific stimulus feature information is integrated with abstract task information in the representations that guide behavior. This insight has motivated theoretical and empirical work examining the close relationship between cognitive control and task representation.